cs.CVNov 10, 2026

Dark Matters: VAE-Aware HDR Encoding for SDR-to-HDR Video Reconstruction

Authors: Donghoon Cha, Bomseok Park, Hansl von Kwon, Ku DoHyung, Hanwool Sul

Organizations: STUDIO FREEWILLUSION Inc. · Seoul National University

Abstract

SDR-to-HDR reconstruction is an ill-posed problem in which clipped highlights, crushed shadows, and quantization artifacts must be recovered while preserving the original content. Latent diffusion models provide powerful generative priors for restoring missing detail, but their effectiveness is constrained by how HDR signals pass through the pretrained VAE. We find that common log-like encodings suppress low-luminance chromatic differences before generation, producing a dark-side collapse in the frozen-VAE latent. We introduce EquivFlow, a compact, invertible exposure-equivariant encoding that maps exposure changes to additive shifts while retaining a scale-independent chromatic component, without modifying the VAE. Encoding-level tests across pretrained image and video VAEs show that its exposure structure and deep-dark chroma transport persist across the tested hosts. Integrated into SDR-to-HDR video diffusion, EquivFlow delivers consistent chromatic-fidelity gains across the tested settings while retaining competitive reconstruction fidelity.

Figures & tables

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. DiffHDR: Re-Exposing LDR Videos with Video Diffusion Models

    Apr 7, 2026Zhengming Yu, Li Ma, Mingming He +11HDR ImagingHDR Video Reconstruction

  2. Single-Shot HDR Recovery via a Video Diffusion Prior

    May 12, 2026Chinmay Talegaonkar, Jinshi He, Christopher McKenna +1Image EnhancementHDR Imaging

  3. X2HDR: HDR Image Generation in a Perceptually Uniform Space

    Feb 4, 2026Ronghuan Wu, Wanchao Su, Kede Ma +2Image GenerationHDR Imaging